Update on statistical power and sample size assessments for cost-effectiveness studies

Abstract

Traditionally, competing healthcare interventions have been compared on their efficacy or effectiveness through clinical trials or epidemiological studies. Computation of sample size and/or statistical power for a proposed study to assess the relative effectiveness of two treatments is central to planning a good study. Economic evaluation studies focus on an additional dimension by comparing these interventions with respect to their cost. With rising healthcare costs and constrained budgets, these studies are increasingly being performed to ascertain which interventions can deliver additional health benefits at a reasonable cost. The design of a cost-effectiveness study for two competing treatments will require assessments of statistical power and sample size in demonstrating both effectiveness and/or cost-effectiveness. If the level of effectiveness of one treatment is known from a clinical trial, the next step is to assess its cost-effectiveness. In other circumstances, an investigator may wish to design a study that simultaneously assesses both effectiveness and cost-effectiveness. Statistical methodologies that have been proposed to address these issues are reviewed.

• confidence interval
• cost-effectiveness plane
• hypothesis tests
• net health cost
• quality-adjusted survival